Few-Shot Cross-Lingual TTS Using Transferable Phoneme Embedding
This addresses the challenge of few-shot cross-lingual TTS for applications requiring rapid adaptation to new languages with minimal data, though it is incremental as it builds on existing transfer learning and phoneme-based methods.
The paper tackles the problem of synthesizing intelligible speech in unseen languages with very limited data by proposing a transferable phoneme embedding framework for cross-lingual text-to-speech. The result shows that using only 4 utterances (about 30 seconds of data) is sufficient for adaptation.
This paper studies a transferable phoneme embedding framework that aims to deal with the cross-lingual text-to-speech (TTS) problem under the few-shot setting. Transfer learning is a common approach when it comes to few-shot learning since training from scratch on few-shot training data is bound to overfit. Still, we find that the naive transfer learning approach fails to adapt to unseen languages under extremely few-shot settings, where less than 8 minutes of data is provided. We deal with the problem by proposing a framework that consists of a phoneme-based TTS model and a codebook module to project phonemes from different languages into a learned latent space. Furthermore, by utilizing phoneme-level averaged self-supervised learned features, we effectively improve the quality of synthesized speeches. Experiments show that using 4 utterances, which is about 30 seconds of data, is enough to synthesize intelligible speech when adapting to an unseen language using our framework.